FISHERY BULLETIN: VOL. 74, NO. 1 



of lat. 33°N and the second two indicating min- 

 utes west of long. 118°W. 



The samples were taken aboard the National 

 Marine Fisheries Service RV David Starr Jordan 

 between 26 July and 3 August 1971. The Shipek 

 bottom sampler was used to obtain the samples of 

 sediment. This device obtains a block of material 

 equal to 400 cm^ of bottom sediment to a depth of 

 about 10 cm, or slightly more, in soft mud or to a 

 depth of half as much or less in coarse sand. 



Two samples were taken at each station in 

 order to obtain an estimate of sampling error. 

 The vessel was allowed to drift while the samples 

 were being taken, so the sample pairs were taken 

 in only approximately the same location. How- 

 ever, agreement in the various parameters be- 

 tween samples from the same station was good. 



The samples were placed in aluminum foil- 

 lined containers of approximately the same size 

 as the sampling bucket and were quick-frozen. 

 They were stored in a freezer until removed for 

 analysis. 



In most samples, DDT was confined to the top 2 

 or 3 cm of the sediment. At most of the stations 

 where the sampler sampled to 10 cm, and at all of 

 the stations where it sampled to a lesser depth, it 

 appeared that all of the DDT under the 400 cm^ 

 had been sampled. In this study, therefore, DDT 

 concentrations are given as the weight of DDT 

 per unit area of bottom to a depth of 10 cm. In a 

 few areas of rapid sedimentation, where the sam- 

 pler sampled to about 10 cm depth, there were 

 still significant amounts of DDT below 10 cm. 

 Estimates for the amounts of DDT below 10 cm 

 are based on core samples taken by other investi- 

 gators in this area. 



The bottom sediment samples were thawed 

 and blended in a 1-gallon Waring^ commercial 

 blender. Before blending, small stones were re- 

 moved from the few samples that contained 

 them. Some samples contained a few small mol- 

 luscs or brittle stars, but these were not removed. 

 Measured amounts of distilled water were added 

 to some of the drier (sandy) samples to facilitate 

 blending. 



A sample of 15 to 20 g of blended sediment was 

 weighed onto a watch glass, dried to constant 

 weight, and reweighed to obtain percent water in 

 the sediment. This gave an index of bottom type 



^Reference to trade name does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



ranging from 30 to 40% water for coarse sand to 

 60 to 70% water for fine silt. 



A second sample weighing about 30 g was 

 weighed into a 1-pint Mason jar for DDT deter- 

 mination. About four or five times as much 

 Na2S04 was weighed into the jar as a drying 

 agent. The sediment and Na2S04 were mixed 

 using a stainless steel spatula, and the mixture 

 was frozen. A cutting assembly was fitted to the 

 jar, and the frozen mixture was thoroughly 

 blended to a powder using an Osterizer blender. 



About 5 g of the powder was weighed into a 

 tared, large disposable pipet (Matheson super 

 pipet) plugged with glass wool. The powder was 

 extracted into a 15-ml graduated centrifuge tube 

 with 5 ml of hexane and 5 ml of acetone. The 

 extract was evaporated to dryness and redis- 

 solved in 1 ml of hexane. This sample was eluted 

 through a super pipet filled with activated alu- 

 mina (McClure 1972) using enough hexane to 

 obtain a 6-ml sample. 



This sample was reduced or increased in vol- 

 ume as required and injected into a model 402 

 Hewlett Packard gas chromatograph (GLC) with 

 a Ni^^ electron capture detector. The 6-foot glass 

 column contained 4% SE-30/6% QF-1 on 100/120 

 mesh Supelcoport. 



There was evidence of a polychlorinated bi- 

 phenyl, Aroclor 1254, in all samples, but the 

 DDT peaks were so dominant in the chromato- 

 grams that they generally obliterated any traces 

 of other chlorinated hydrocarbons within their 

 range. Only the six peaks representing the ortho- 

 para and para-para forms of DDE, DDD, and 

 DDT were quantified. "Total DDT" is used to 

 designate the sum of these six analogs. 



RESULTS AND DISCUSSION 



Fifty-five correlations were obtained for 11 

 parameters to determine various DDT relation- 

 ships. The 55 correlations were obtained for all 

 103 stations (Table 1, values above 1.000 correla- 

 tion diagonal) and for 76 stations leaving out 

 those 27 stations having total DDT readings 

 greater than 100 mg/m^ (Table 1, values below 

 1.000 correlation diagonal). For 100 observations 

 a correlation coefficient of 0.254 indicates a 

 probability of 0.01. Logarithms were used for 

 total DDT and distance from outfall, arithmetic 

 values for the other nine measurements. 



There is a very high negative correlation be- 

 tween log total DDT and log distance from the 



28 



